PROJECT SUMMARY/ABSTRACT Gastrointestinal Stromal Tumor (GIST) is the most common sarcoma in the U.S. The treatment of GIST with imatinib, a tyrosine kinase inhibitor (TKI), that targets the KIT oncogene, provides proof of principle for precision oncology. We now understand that GISTs also occur due to mutations in PDGFRA, RAS/BRAF/NF1, SDHA-D, as well as FGFR1 and ETV6-NTRK3 fusions. Thus, there are several GIST subtypes and many of these are imatinib resistant. Imatinib markedly improves patient survival, but fails to cure GIST as many patients develop recurrences ?8-12 months after stopping imatinib. These recurrences raise the question as to whether GIST persistence post-imatinib occurs via an unappreciated mechanism. We hypothesize that KITneg tumor cells, which lack KIT protein expression, represent TKI-resistant GIST cancer stem cells (CSCs) based on our preliminary data that: 1) human KIT mutant GIST cell lines possess KITneg cells that self-renew, form tumorspheres, and have primary TKI-resistance; 2) both KIT and non-KIT mutant resected human GISTs possess KITneg cells and have increased expression of stem cell genes following TKI therapy; and 3) these KITneg cells overexpress druggable targets (i.e., Hedgehog, AMPK, NF-kB, and AXL) involved in CSC biology. In this project, we propose to further characterize and target these KITneg GIST CSCs to overcome this previously unappreciated, and clinically relevant, source of TKI-resistance using state-of-the-art experimental techniques and advanced computational analysis methods. In Aim 1, we will define the molecular and functional properties of KITneg cells to determine how they promote GIST growth and TKI resistance. We will determine the self-renewal and differentiation potential of each putative CSC population following serial TKI treatments that parallel FDA- approved therapies. We will then identify conserved proteins and pathways of isolated TKI-resistant KITneg and KIThi cells. We will also determine common mechanisms of TKI-resistance amongst KITneg cells following serial therapies. In Aim 2, we will characterize the CSC properties and molecular heterogeneity of KITneg cells from primary KIT mutant and non-KIT mutant tumors. We will determine the inter-/intra-tumoral self-renewal, differentiation, and tumorsphere formation for primary putative CSC populations from GISTs. Thus, we will determine conserved pathways in putative CSCs from genomically diverse GISTs. In Aim 3, we will validate druggable targets (e.g., Hedgehog, AMPK, NF-kB, and AXL) and exploit these pathways for eradication of KITneg cells in vitro and in vivo. We will validate our findings with genetic approaches to determine if these targets are required to maintain viability. This project will explore the commonalities and differences across TKI-treated and genomically diverse GIST CSCs, which represent a novel target for overcoming disease persistence and TKI- resistance. Our proposal will identify novel drug targets and agents for killing GIST CSCs, and thus will have immediate and significant impact on clinical trial design of new agent(s) alone or in combination with imatinib for curing GIST.